When information comparing mortality and other complication rates among the hospitals of a state or region is made available to physicians, will the quality of care be improved? This is a feasibility study to plan an experiment to answer this question. The overall goal of the feedback experiment is the design, implementation and evaluation of a quality control program that uses readily available data such as hospital discharge data sets (HDDS) and health insurance data to monitor and improve patient care outcomes. The quality objectives of the experiment are two-fold: (1) to improve the understanding among practicing physicians of the risks and benefits of various treatments and thus to influence overall practice; (2) to improve the quality of care in specific hospitals where the outcomes for certain categories of its patients are significantly worse. The specific aims of this feasibility study are to (1) develop statistical criteria for identifying hospitals with worse than average performance; (2) design and pre-test reports that convey information on the risks of using spcific treatments as well as how average risks compare with that of specific hospitals; (3) design the experiment itself; (4) obtain the active support of a state medical society and/or the Professional Review Organization in state as sponsors of the experiment.